Metformin influences drug sensitivity in pancreatic cancer cells

16 

Loading....

Loading....

Loading....

Loading....

Loading....

Full text

(1)

UNCORRECTED

PROOF

Contents lists available at ScienceDirect

Advances in Biological Regulation

journal homepage: www.elsevier.com

Metformin influences drug sensitivity in pancreatic cancer cells

Saverio Candido

⁠a⁠, ⁠1

, Stephen L. Abrams

⁠b⁠, ⁠1

, Linda Steelman

⁠b

, Kvin Lertpiriyapong

⁠c⁠, ⁠2

,

Alberto M. Martelli

⁠d

, Lucio Cocco

⁠d

, Stefano Ratti

⁠d

, Matilde Y. Follo

⁠d

, Ramiro M. Murata

⁠b⁠, ⁠e

,

Pedro L. Rosalen

⁠f

, Paolo Lombardi

⁠g

, Giuseppe Montalto

⁠h⁠, ⁠i

, Melchiorre Cervello

⁠i

, Agnieszka Gizak

⁠j

,

Dariusz Rakus

⁠j

, Pann-Gill Suh

⁠k

, Massimo Libra

⁠a

, James A. McCubrey

⁠b⁠, ⁠∗

aDepartment of Biomedical and Biotechnological Sciences - Pathology & Oncology Section, University of Catania, Catania, Italy bDepartment of Microbiology & Immunology, Brody School of Medicine, East Carolina University, Greenville, NC 27834, USA cDepartment of Comparative Medicine, Brody School of Medicine at East Carolina University, USA

dDepartment of Biomedical and Neuromotor Sciences, Università di Bologna, Bologna, Italy eDepartment of Foundational Sciences, School of Dental Medicine, East Carolina University, USA

fDepartment of Physiological Sciences, Piracicaba Dental School, State University of Campinas, Piracicaba, Brazil gNaxospharma, Via Giuseppe Di Vittorio 70, Novate Milanese 20026, Italy

hBiomedical Department of Internal Medicine and Specialties, University of Palermo, Palermo, Italy

iConsiglio Nazionale delle Ricerche, Istituto di Biomedicina e Immunologia Molecolare “Alberto Monroy”, Palermo, Italy jDepartment of Molecular Physiology and Neurobiology, Wroclaw University, Wroclaw, Poland

kDepartment of Biological Sciences, Ulsan National Institute of Science and Technology (UNIST), Ulsan, Republic of Korea

A R T I C L E I N F O

Keywords: Metformin AMPK Rapamycin mTORC1

Signal transduction inhibitors

A B S T R A C T

Pancreatic ductal adenocarcinoma (PDAC) is an aggressive, highly metastatic malignancy and ac-counts for 85% of pancreatic cancers. PDAC patients have poor prognosis with a five-year survival of only 5–10% after diagnosis and treatment. Pancreatic cancer has been associated with type II diabetes as the frequency of recently diagnosed diabetics that develop pancreatic cancer within a 10-year period of initial diagnosis of diabetes in increased in comparison to non-diabetic pa-tients. Metformin is a very frequently prescribed drug used to treat type II diabetes. Metformin acts in part by stimulating AMP-kinase (AMPK) and results in the suppression of mTORC1 activity and the induction of autophagy. In the following studies, we have examined the effects of met-formin in the presence of various chemotherapeutic drugs, signal transduction inhibitors and nat-ural products on the growth of three different PDAC lines. Metformin, by itself, was not effective at suppressing growth of the pancreatic cancer cell lines at concentration less than 1000nM, how-ever, in certain PDAC lines, a suboptimal dose of metformin (250nM) potentiated the effects of various chemotherapeutic drugs used to treat pancreatic cancer (e.g., gemcitabine, cisplatin, 5-flu-orouracil) and other cancer types (e.g., doxorubicin, docetaxel). Furthermore, metformin could in-crease anti-proliferative effects of mTORC1 and PI3K/mTOR inhibitors as well as natural products such as berberine and the anti-malarial drug chloroquine in certain PDAC lines. Thus, metformin can enhance the effects of certain drugs and signal transduction inhibitors which are used to treat pancreatic and various other cancers.

Corresponding author.

Email address: mccubreyj@ecu.edu (J.A. McCubrey)

1Co-first authors, these authors contributed equally to these studies.

2Current address: Center of Comparative Medicine and Pathology, Memorial Sloan-Kettering Cancer Center, Weill Cornell Medicine and the Hospital for Special

Surgery, New York City, New York, USA. https://doi.org/10.1016/j.jbior.2018.02.002

Received 30 January 2018; Received in revised form 3 February 2018; Accepted 5 February 2018 Available online xxx

(2)

UNCORRECTED

PROOF

1. Introduction

Pancreatic cancer is a devastating cancer worldwide with no truly effective cures. Surgery and some forms of chemotherapy are

clinical options however, the mean 5-year survival is less than 5–10%. By the time that pancreatic ductal adenocarcinoma (PDAC)

is diagnosed, PDAC has often spread to distant sites which makes therapy difficult. The increasing incidence and mortality from

PDAC makes attempts to cure this disease of paramount importance to people worldwide. Treatment combining surgical resection

and chemotherapy is only minimally effective (Siegel et al., 2013). PDAC is strongly associated with environmental factors such as:

cigarette smoke, diet, diabetes, obesity, alcohol consumption, gender and low physical activity (Iodice et al., 2008; Arslan et al., 2010;

Jiao et al., 2010; Michaud et al., 2010; Li, 2012; Lucenteforte et al., 2012; Maisonnneuve et al., 2012; Bosetti et al., 2014).

The association between type II diabetes chronic inflammation and pancreatic cancer has also been observed in mouse models.

In this mouse model, the type-II diabetes drug metformin could suppress cancer cell proliferation in the mice. Chronic inflammation

was observed to have only minor effects on the pathophysiology of the established adenocarcinoma in this orthotopic pancreatic

transplant model (Zechner et al., 2015). Type II diabetes may be the third highest risk factor for pancreatic cancer, behind cigarette

smoking and obesity (Lowenfels and Maisonneuve, 2006; Maisonneuve and Lowenfels, 2015).

PDAC is also associated with mutations at many oncogenes/tumor suppressor genes. Some of the most frequently altered genes are:

KRAS, TP53, CDKN2A, and SMAD4 (Rasheed et al., 2010; Klein, 2012; Fitzgerald et al., 2015; McCubrey et al., 2015). Approximately

90% of pancreatic cancers contain mutations in KRAS. These mutations result in an activated KRAS oncoprotein which interacts with

downstream signaling pathways leading to continuous cell growth, prevention of apoptosis, senescence, and chemotherapeutic drug

resistance (Adamska et al., 2017).

KRAS is often normally activated by various growth factor receptors and is part of the RAS/RAF/ERK, PI3K/PTEN/AKT/mTORC1/

GSK-3 and other interacting signaling pathways. The roles of these and other signaling pathways in various disorders, cancer and

development have been recently reviewed (McCubrey et al., 2014a, 2014b; Chappell et al., 2016; Ghim et al., 2016; Giudici et al.,

2016; Kang et al., 2016; McCubrey et al., 2016; Yang et al., 2016; Cervello et al., 2017; Jhanwar-Uniyal et al., 2017; McCubrey et al.,

2017a, 2017b, 2017c, 2017d, 2017e; Pappas et al., 2017; Ruvolo, 2017; Ryuno et al., 2017). As KRAS has been frequently observed

to be mutated in pancreatic cancer, attempts have been developed to generate effective RAS inhibitors as well as additional signaling

molecules in this pathway. Attempts to inhibit the RAS/PI3K/PTEN/Akt/mTORC1/GSK-3 and other signaling pathways have been a

central focus in many basic science laboratories and pharmaceutical companies for over three decades due to the involvement

onco-genes in this pathway in human cancers (Yang et al., 2013; Fitzgerald et al., 2015; McCubrey et al., 2015; Anderson et al., 2016;

Banfic et al., 2016; Cocco et al., 2016; Erneux et al., 2016; Falasca and Ferro, 2016; Fields et al., 2016; Geck and Toker, 2016; Maczis

et al., 2016; Perdios et al., 2016; Pyne et al., 2016; Scarlata et al., 2016; Scoumanne et al., 2016; Tanaka et al., 2016; Ando et al.,

2017; Carman and Han, 2017; Fujisawa, 2017; Geffken and Spiegel, 2017; Guo et al., 2017; Jang et al., 2017; Leonard and Johnson,

2017; Liu et al., 2017; Matsuzawa, 2017; McCubrey et al., 2017a, 2017b, 2017c, 2017d; Obsil and Obsilova, 2017; Okazaki, 2017;

Pyne et al., 2017; Ramos et al., 2017; Ratti et al., 2017; Rebello et al., 2017; Roth and Frohman, 2017; Rusnak and Fu, 2017; Sakane

et al., 2017; Scarlata et al., 2017; Vasquez et al., 2017). Multiple signaling pathways may be affected by mutant oncogenes and

in-flammatory signals as key molecules of one signaling pathway may interact with other important pathways (Coant et al., 2017; Campa

and Hirsch, 2017; Ebenezer et al., 2017; Gowda et al., 2017a, 2017b; Hatch et al., 2017; Hermida et al., 2017; Mérida et al., 2017;

Nishida et al., 2017; Ricciardi et al., 2017; Ruvolo, 2017; Ruzzene et al., 2017; Schrock et al., 2017; Shears et al., 2017; Steelman et

al., 2017; Ramazzotti et al., 2017; Yamauchi et al., 2017).

1.1. Metformin regulates key signaling pathways in diabetes which may have effects on pancreatic cancer

Metformin (N,N-Dimethylimidodicarbonimidic diamide) is a type II diabetes drug which may also have effects on cancer stem cells

(CSCs) (Lonardo et al., 2013; Liu et al., 2016). Metformin can activate 5′ AMPK and suppress mTOR activity (Rozengurt et al., 2014)

(See Fig. 1). The PI3K/PTEN/Akt/mTORC1 and Raf/MEK/ERK pathways are inhibited when AMPK is activated. These pathways often

serve to regulate the translation of mRNAs which have 3'regions which make them difficult to translate (Leibovitch and Topisirovic,

2017).

The effects of metformin on PDAC development and survival remain controversial. Some studies have demonstrated that

met-formin may improve survival in some PDAC patients which had undergone surgical resection (Cerullo et al., 2017). Another study

observed a survival benefit in diabetes patients who used metformin before the development of PDAC (Amin et al., 2016).

However, other studies have not observed statistically significant better survival in patients who used metformin and had surgery

and have suggested that the statistical methods of patient selection were not appropriate in other studies (Chaiteerakij et al., 2016).

An additional randomized phase II trial with approximately 60 patients did not demonstrate a benefit with metformin consumption

to a standard systemic therapy with cisplatin, epirubicin, capecitabine and gemcitabine patients with metastatic pancreatic cancer.

The authors concluded that addition of the metformin dose used to treat type II diabetics did not improve outcome of patients with

metastatic pancreatic cancer (Reni et al., 2016). A large study (907 patients) using data from the Netherlands Comprehensive

Can-cer Center observational cohort (1998–2011) also indicated that there was no association in terms of survival in patients using met

(3)

UNCORRECTED

PROOF

Fig. 1. Overview of Effects of Metformin on Signaling Pathways in Pancreatic Cancer. In this figure, the KRAS/PI3K/PTEN/AKT/mTORC pathway as KRAS is

frequently mutated in pancreatic cancer is shown. The effects of metformin on the induction of AMPK and rapamycin on suppression of mTORC1 and induction of au-tophagy are indicated. In addition, the effects of chloroquine on the suppression of auau-tophagy are indicated. Some sites where effective inhibitors have been developed to either suppress or activate key signaling molecules are indicated.

formin in PDAC patients (Frouws et al., 2017). Finally, some basic research studies with patient derived xenografts (PDX) in mice

have indicated that metformin treatment did not inhibit the growth of the PDX in the four different PDX mouse models utilized in

that study. The authors proposed that the PDAC cells may have become resistant to metformin by various mechanisms including

lack of activation of AMPK or reactivation of mTOR (Lipner et al., 2016). Thus, while some clinicians, epidemiologists and basic

scientists have concluded that metformin usage does not improve the survival of PDAC patients, there remains some open

ques-tions as to whether metformin can enhance the effects of other drugs which may be used in PDAC and other types of cancers.

There are at least 16 clinical trials with the anti-type II diabetes drug metformin and pancreatic cancer https://clinicaltrials.gov/ct2/

results?cond=pancreatic+cancer&term=metformin&cntry=&state=&city=&dist=]. Some of these clinical trials are examining the

effects of combining metformin and chemotherapeutic drugs, rapalogs and other drugs. These trials differ in their levels of progress

and some have been discontinued.

(4)

UNCORRECTED

PROOF

1.2. Signaling pathways regulated by metformin-key roles in many physiological processes and cancer development

The LKB1/AMP activated protein kinase (LMPK) pathway is a key pathway investigated initially in metabolic diseases such as

diabetes (Steelman et al., 2016). However, given the interrelationship between metabolism and cancer, the LMPK pathway is now

rec-ognized to play important roles in cancer and other diseases (Martelli et al., 2012). This pathway can be activated by the commonly

prescribed drugs, such as N,N-dimethylimidodicarbonimidic diamide (metformin) and 5-aminoimidazole-4-carboxamide

1-β-D-ribo-furanoside (AICAR). AICAR is prescribed to reduce cardiac ischemic injury as well as diabetes care. LKB1/AMPK activation can result

in cell cycle arrest, caspase-dependent apoptosis or autophagy. mTORC1-controlled protein translation is suppressed by metformin.

Many of the targeted proteins are critical in the growth of malignant cells. Interestingly, the translation of these proteins is not as well

inhibited with allosteric mTORC1 inhibitors, such as rapamycin and its derivatives but their translation may be better suppressed by

metformin. Thus, the LKB1/AMPK pathway is key in regulating diseases such as diabetes as well as the proliferation and survival of

cancer cells. Drugs targeting and activating LKB1/AMPK may be a less toxic and cost-effective treatment option for certain

malignan-cies (Bressanin et al., 2012; McCubrey et al., 2014b).

LKB1 is considered a tumor suppressor gene. Gatekeeper mutations of LKB1 are responsible for the rare Peutz-Jeghers Syndrome

(PJS). PJS is a cancer-prone syndrome (Jansen et al., 2009). LKB1 is a critical regulator of energy/metabolism control as well as

po-larity. LKB1 exerts it vast effects via diverse effectors (Godlewski et al., 2010; van der Velden and Haramis, 2011). AMPK functions as

a metabolic gatekeeper important in many diseases including diabetes, cancer and neurologic disorders. Inhibiting mTORC1 activity

by drugs such as metformin and other drugs may not only aid in the treatment of diabetics, but also improve cancer therapies and

increase longevity (Liu et al., 2011; Vazquez-Martin et al., 2011; Mashhedi et al., 2011; Oliveras-Ferraros et al., 2011). Rapamycin

may also have some similar but not identical effects.

1.3. Pancreatic cancer stem cells (CSCs) and potential targeting approaches

The concept of CSCs has emerged in different types of cancers including PDAC (Fitzgerald and McCubrey, 2014; Fitzgerald et al.,

2015; Xu et al., 2015; Steelman et al., 2016; McCubrey et al., 2017c). The PDAC CSCs contain certain stem-like properties. These

properties include: the expression of certain cell surface antigens, the increased expression of proteins that are involved in

chemo-ther-apeutic drug resistance (drug pumps) and expression of certain transcription factors.

More recently, the expression of various micro RNAs (miRs) have been associated with CSCs (Steelman et al., 2016; McCubrey et

al., 2017c). Attempts to target CSCs with various drugs and other methods has emerged as the CSCs are thought to be responsible for

cancer initiation, drug resistance and reemergence after various therapeutic approaches.

The CSCs often may differ from the bulk of the tumor cells (BC) present in the cancer patient as the CSCs have properties of stem

cells. CSCs may be obtained from certain established PDAC lines based on the expression of certain cell surface markers and their

ability to forms spheroids in culture. CSCs have increased expression of CD44, CD24, CD133, aldehyde dehydrogenase-1 (ALDH1),

epithelial-specific antigen (ESA), activin receptor type-1B (ALK4), CXCR4, CXCL12 and Notch compared to the BCs (Fitzgerald and

McCubrey, 2014; Fitzgerald et al., 2015).

Often cells are characterized by their growth properties. The adherent cells (Adh.) appear to differ from spheroid forming cells.

Spheroid forming cells may have some characteristic of CSC. The spheroid forming cells is more of a morphological criterion. In the

following studies, we have focused only on Adh. cells as we have compared three different PDAC lines which may have different

growth properties and differential abilities to generate spheroid cells.

The KRAS-JNK signaling axis is also important in the self-renewal and tumor initiating properties of pancreatic CSCs. Suppression

of KRAS activity resulted in inhibition of the JNK pathway and negatively affected the pancreatic CSC population in pancreatic cancer

pre-established xenograft models (Okada et al., 2014).

Upregulation of the hedgehog [HH] pathway was observed in drug-resistant BxPC-3 and Panc3.27 pancreatic cells. The slower

cycling pancreatic cells were more invasive and had a higher tumorigenic potential compared to the faster cycling cells. The slower

cycling cells had morphological properties consistent with cells which had undergone epithelial mesenchymal transition (EMT)

(Dembinski and Krauss, 2009). HH inhibitors may be effective in targeting pancreatic CSCs if they are combined with other drugs/

inhibitors (Sandhiya et al., 2013).

The natural product cyclopamine targets smoothened (SMO) which is an important component in the HH pathway. Vismodegib is

a HH inhibitor developed by Genetech (LoRusso et al., 2011). It also targets smoothly SMO and results in the inactivation of GLI1 and

GLI2 transcription factors. Vismodegib is FDA approved to treat basal cell carcinoma patients and is also being evaluated with

pancre-atic, colorectal and other cancer patients (Sekulic et al., 2012, 2017; Sandhiya et al., 2013). However, a clinical trial with combined

gemcitabine and vismodegib treatment in PDAC did not show an enhancement of survival in an unselected cohort (Catenacci et al.,

2015).

Suppression of either mTOR by rapamycin or HH by cyclopamine was not by themselves sufficient to effectively eliminate the

pancreatic CSC pool. In contrast, combining mTORC1 and HH inhibitors suppressed the pancreatic CSCs (Mueller et al., 2009).

Inter-estingly, rapamycin reduced the stemness properties (viability and sphere formation ability) of CD133 pancreatic cancer cells while

(5)

UNCORRECTED

PROOF

suppression of mTORC1 and inhibition of HH activity had different effects on pancreatic cancer stem-like cells (Matsubara et al.,

2013).

Metformin treatment was observed to have negative effects on cell survival, clonogenicity, sphere forming capacity in both

gemc-itabine-sensitive and resistant PDAC. Metformin inhibited the expression of CSC markers, CD44, EpCAM, enhancer of zeste homolog

2 (EZH2), Notch-1, Nanog and Oct4. In contrast, metformin induced the expression of various miRs including: let-7a, let-7b, miR-26a,

miR-101, miR-200b, and miR-200c. The expression of these miRs are often detected at different levels in pancreatic cancer Adr. cells

and pancreatic sphere colonies (Bao et al., 2012, 2014).

In spheroid forming PDAC models, a different pattern of miR expression was observed consisting of miR-99a, miR-100, miR-125b,

miR-192, and miR-429 than in Adh. cells (Jung et al., 2011). Holoclone forming cells were isolated from pancreatic cancer

BxPC-3cells that were enriched in PDAC CSCs. Proteins such as CXCR4, Polycomb complex protein (BMI1), GLI1, and GLI were

ex-pressed in the holoclones. In addition, miR-214, miR-21, miR221, miR-222 and miR-155 were observed (Tan et al., 2011). These

results indicate that treatment of PDAC cells with metformin may alter the expression of various miRs which have different targets.

Certain drugs may target adherent and spheroid cells differently. Drugs which affect metabolism such as rapamycin and metformin

may have more effects on spheroid than adherent cells. The anti-malarial drug chloroquine has effects on pancreatic spheroid CSCs

(Balic et al., 2014a, 2014b). Chloroquine has effects on the C-X-C chemokine receptor type 4 (CXCR4) and internalizes it. This makes

cells less responsive to chemokine (C-X-C motif) ligand 12 (CXCL12). HH signaling is decreased and there are lower levels of PTCH

and SMO on the cell surface. In addition, chloroquine effects the levels of GLI and Cyclin D1 proteins in the cytoplasm and suppresses

ERK and STAT5 activation (Balic et al., 2014a).

1.4. Catabolic sources of energy and metabolic reprogramming in CSCs-effects of metformin

Breast and other types of CSCs have been observed to have enhanced ability to utilize catabolic sources of energy under starvation

circumstances. Metabolic reprogramming may be an adaptive strategy by which CSCs survive during nutrient deprivation. Metabolic

reprogramming during cancer progression is important (Cuyas et al., 2014). Other models documenting the importance of glycolysis

and tumor metabolism have been presented (Alfarouk et al., 2014).

Elevated glucose uptake is a manifestation present in many metastatic tumors. The expression of mutant oncogenes can contribute

to this phenomenon. RAS pathway activation increases glycolytic flux into lactate. The hepatocyte growth factor (HGF) binds the MET

tyrosine kinase receptor. This can result in activation of RAS and downstream ERK. This leads to increased motility and glucose

up-take. This may regulate the metabolism of certain breast and other cancer types as well as influence blood-flow in the tumors (Natan

et al., 2014).

The effects of metformin and the related biguanide phenformin were examined in an SRC-inducible model of transformation as

well as in mammosphere-derived breast CSCs by metabolomics to evaluate how these drugs might preferentially effect CSCs.

Dur-ing the process of cellular transformation, treatment with the biguanides suppressed the increase of glycolytic intermediates and

decreased tricarboxylic acid (TCA) intermediates at certain stages of transformation. In contrast, in the breast CSCs, the effects on

glycolytic and TCA intermediates were not as pronounced but reduction in nucleotide triphosphates levels were documented which

could suppress nucleotide synthesis. Thus, the biguanides inhibited mitochondrial complex 1, while their metabolic effects differed at

various stages of the transformation process (Janzer et al., 2014).

Metformin was proposed to induce metabolic reprogramming of chemoresistant breast cells. Metformin stimulated metabolism

in ALDH-bright but not ALDH-low cells. Metformin treatment altered glutathione metabolism in the ALDH-bright cells in

compari-son to the ALDH-low cells. As in other studies, metformin altered miR expression in ALDH-bright cells. Interestingly, many of the

affected miRs were believed to target metabolic pathways. Thus metformin may reduce the differences between the chemoresistant

ALDH-bright and chemosensitive ALDH-low cells (Cioce et al., 2014).

Metformin-resistance in breast and other cancer types may result in cells with a metastatic stem-like phenotype. Resistance to

metformin lead to expression of many genes encoding: migration and invasion factors, stem cell markers, pro-metastatic lipases and

components of the degradome. Downregulation of genes involved in the G

⁠2

/M DNA damage checkpoint regulation that maintain

genome stability as well as alteration of gene expression with pro-autophagic features were also observed (Oliveras-Ferraros et al.,

2014; Menendez et al., 2014).

The effects of metformin and hyperthermia were examined on MIA-PaCa-2 pancreatic and MCF-7 and MDA-MB-231 breast

can-cer cells. Metformin was determined to be preferentially cytotoxic to CD44high/CD24low MCF-7cells and CD44high/CD24high

Mia-PaCa-2cells. The CD44high/CD24low and CD44high/CD24high phenotypes are associated with breast and pancreatic CSCs

re-spectively. Exposure of the cells to 42°C for 1h enhanced the ability of metformin to kill BC cancer cells as well as CSCs. Hyperthermia

was determined to activate AMPK and inactivate mTORC1 and downstream p70S6K. Thus, hyperthermia could enhance the cytotoxic

effects of metformin on both breast and pancreatic CSC in this experimental model system (Lee et al., 2014).

Metformin has effects on the expression of genes involved in metabolism in breast and other cancer cell types. Metformin

sup-pressed fatty acid synthetase (FASN) in triple negative breast cancer (TNBC) by inducing the expression miR-193b. miR-193b bound

the 3′UTR of FASN. Importantly, the effects of miR-193b were observed in TNBC but not in non-transformed breast epithelial cells.

Metformin was shown to induce miR-193 which reduced mammosphere formation in TNBC lines (Wahdan-Alaswad et al., 2014).

Metformin may be synthetically lethal with glucose withdrawal in breast and other types of CSCs. CSCs have increased reliance

on Warburg-like aerobic glycolysis to maintain CSC stemness and immortality. These stemness properties are lost upon metformin

(6)

UNCORRECTED

PROOF

treatment. Metformin inhibited the ability of oncogenes such as HER2 to prevent apoptosis under conditions of glucose-deprivation

of HER2-positive breast CSCs (Menendez et al., 2012).

2. Results

Given the various documented effects of metformin on various types of cancer cells and the different potential mechanisms by

which it could induce these different activities, we examined the abilities of metformin to suppress growth in three different

pancre-atic cancer cell lines ASPC-1, BxPC-3 and MIA-PaCa-2. As observed in Fig. 2, treatment with metformin by itself with concentrations

up to 5000nM did not elicit growth inhibitory effects on BxPC-3 or MIA-PaCa-2cells as IC

⁠50

values were not obtained. Treatment of

ASPC-1cells with 5000nM metformin did result in an IC

⁠50

. It is not surprising that metformin is not strongly toxic on cells as it is

used for daily treatment of certain type II diabetes patients. The metformin levels in patient's serum taking this drug is approximately

1–20μM. The levels of metformin in the serum depends on the kidney glomerular filtration rate (GFR) activity levels (Frid et al.,

2010).

2.1. Effects of combining a suboptimal dose of metformin with drugs used to treat pancreatic cancer patients

The effects of combining a suboptimal dose of metformin (250nM) with chemotherapeutic drugs used to treat pancreatic cancer

patients and the HH inhibitor visomedigib were examined in Figs. 3 and 4. The experiments in the individual panels were both

per-formed on the same day and most of these experiments in the two figures were perper-formed the same day. Addition of a suboptimal

dose of metformin lowered the gemcitabine IC

⁠50

from 700nM to 25nM (28-fold) in ASPC-1cells (Fig. 3, Panel A). Addition of a

sub-optimal dose of metformin lowered the gemcitabine IC

⁠50

from 130nM to 9nM (14.4-fold) in BxPC-3cells (Fig. 3, Panel B). Addition of

a suboptimal dose of metformin lowered the gemcitabine IC

⁠50

from 9nM to 3nM (3-fold) in MIA-PaCa-2cells (Fig. 3, Panel C). These

experiments demonstrated that suboptimal doses of metformin could lower the concentration of gemcitabine required to reach the

IC

⁠50

for three different PDAC lines. Furthermore, these experiments document the differential sensitivity of these three PDAC lines to

gemcitabine, almost a 100-fold difference in sensitivity between MIA-PaCa-2cells and ASPC-1cells.

Addition of a suboptimal dose of metformin lowered the 5-fluorouracil IC

⁠50

from 9μM to 0.25μM (36-fold) in ASPC-1cells (Fig. 3,

Panel D). Addition of a suboptimal dose of metformin lowered the 5-fluorouracil IC

⁠50

from 3.5nM to 0.5nM (7-fold) in BxPC-3cells

(Fig. 3, Panel E). Addition of a suboptimal dose of metformin lowered the 5-fluorouracil IC

⁠50

from 5μM to 0.6μM (3-fold) in

MIA-PaCa-2cells (Fig. 3, Panel F).

Addition of a suboptimal dose of metformin lowered the cisplatin IC

⁠50

from 15μM to 0.7μM (21.4-fold) in ASPC-1cells (Fig.

4, Panel A). Addition of a suboptimal dose of metformin lowered the cisplatin IC

⁠50

from approximately 200μM–1μM (200-fold) in

BxPC-3cells (Fig. 4, Panel B). Addition of a suboptimal dose of metformin did not lower the cisplatin IC

⁠50

in MIA-PaCa-2cells (Fig. 4,

Panel C). These experiments demonstrated that suboptimal doses of metformin could lower the concentration of cisplatin required to

reach the IC

⁠50

for two different, ASPC-1 and BxPC-3 PDAC lines but not in MIA-PaCa-2cells.

Vismodegib is a small molecule inhibitor which targets HH signaling and is being used to treat basal cell carcinoma (Sekulic et

al., 2012, 2017). It is also being evaluated in other cancers such as PDAC (Catenacci et al., 2015). Addition of a suboptimal dose

of metformin did not lower the vismodegib IC

⁠50

in either ASPC-1 or BxPC-3cells (Fig. 4, Panels D and E), however, it did decrease

growth observed in BxPC-3cells at high concentrations of vismodegib (Fig. 4, Panel E). It also decreased growth and reduced the IC

⁠50

in MIA-PaCa-2cells (Panel F).

Fig. 2. Effects of Metformin on the Proliferation of Three Pancreatic Cancer Cell Lines. ASPC-1 (red squares), BxPC-3 (green diamonds) and MIA-PaCa-2 (blue

circles) and MCF-7 (red circles) cells were titrated with different concentrations of metformin for 4 days before MTT analysis was performed. Arrows on the X-axis indicate where the IC⁠50s can be estimated. These experiments were repeated 3 times and similar results were obtained. Adh.=adherent cells. (For interpretation of the

(7)

UNCORRECTED

PROOF

Fig. 3. Effects of Gemcitabine and 5-Fluorouracil in the Absence and Presence of Suboptimal Doses of Metformin on the Proliferation of Three Pancreatic Cancer Cell Lines. ASPC-1 (Panel A), BxPC-3 (Panel B) and MIA-PaCa-2 (Panel C) were titrated with different concentrations of gemcitabine in the absence (red

squares) and presence (blue triangles) of suboptimal doses of 250nM metformin. Arrows on the X-axis indicate where the IC⁠50s can be estimated. In Panel A, the

two-tailed P value is less than 0.0001 between gemcitabine and gemcitabine and metformin-treated ASPC-1cells and considered to be extremely statistically significant. In Panel B, the two-tailed P value is less than 0.0001 between the gemcitabine and the gemcitabine and metformin-treated BxPC-1cells and considered to be extremely statistically significant. In Panel C, the two-tailed P value is less than 0.0001 between the gemcitabine and the gemcitabine and metformin-treated MIA-PaCa-2cells are less than 0.0001 and considered to be extremely statistically significant. ASPC-1 (Panel D), BxPC-3 (Panel E) and MIA-PaCa-2 (Panel F) were titrated with different concentrations of 5-fluorouracil in the absence (red squares) and presence (blue triangles) of suboptimal doses of 250nM metformin. Arrows on the X-axis indicate where the IC⁠50s can be estimated. In Panel D, the two-tailed P value equals 0.0002 between the 5-fluorouracil and the 5-fluorouracil and metformin-treated ASPC-1cells

and considered to be extremely statistically significant. In Panel E, the two-tailed P value is less than 0.0001 between the 5-fluorouracil and the 5-fluorouracil and met-formin-treated BxPC-3cells is less than 0.0001 and considered to be extremely statistically significant. In Panel F, the two-tailed P value is less than 0.0001 between the 5-fluorouracil and the 5-fluorouracil and metformin-treated MIA-PaCa-2cells and considered to be extremely statistically significant. These experiments were repeated 3 times and similar results were obtained. Adh.=adherent cells. (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.)

2.2. Effects of combining a suboptimal dose of metformin with commonly used chemotherapeutic drugs

The effects of metformin on two commonly used chemotherapeutic drugs (doxorubicin and docetaxel) were examined (Fig. 5).

Addition of a suboptimal dose of metformin lowered the doxorubicin IC

⁠50

from 130nM to 6μM (21.7-fold) in ASPC-1cells (Fig. 5,

Panel A). Addition of a suboptimal dose of metformin lowered the growth of BxPC-3cells when cultured in doxorubicin up to

approx-imately 400nM, but did not lower the doxorubicin IC

⁠50

in BxPC-3cells which was approximately 600nM (Fig. 5, Panel B). Addition

of a suboptimal dose of metformin lowered the doxorubicin IC

⁠50

in MIA-PaCa-2cells from approximately 90nM–45nM (about 2-fold)

(Fig. 5, Panel C). These experiments demonstrated that suboptimal doses of metformin could lower the concentration of doxorubicin

required to reach the IC

⁠50

for two different PDAC lines, ASPC-1 and MIA-PaCa-2 lines but not in BxPC3 cells. These experiments also

demonstrated that some PDAC lines such as BxPC3 are more resistant to doxorubicin than ASPC-1 and MIA-PaCa-2cells.

Addition of a suboptimal dose of metformin lowered the docetaxel IC

⁠50

from 6.5nM to 0.2nM (3.3-fold) in ASPC-1cells (Fig. 5,

Panel D). Addition of a suboptimal dose of metformin did not lower the docetaxel IC

⁠50

in BxPC-3cells (Fig. 5, Panel E). If anything, the

suboptimal dose of docetaxel inhibited the suppressive effects of docetaxel in BxPC-3cells Addition of a suboptimal dose of metformin

did not lower the docetaxel IC

⁠50

in MIA-PaCa-2cells (Fig. 5, Panel F).

2.3. Effects of combining a suboptimal dose of metformin with drugs used with PI3K/PTEN/Akt/mTORC pathway inhibitors

Most PDACs have mutations in KRAS which result in increased activation (Rasheed et al., 2010; Klein, 2012; Fitzgerald et al.,

2015; McCubrey et al., 2015). KRAS can be an upstream activator of PI3K. Thus, the targeting of KRAS, PI3K and mTORC1 has been

an active area in basic pancreatic cancer research. One of the effects of metformin is the suppression of mTORC1 activity which

leads to the induction of autophagy. Likewise, the mTORC1 blocker and the dual PI3K/mTOR inhibitor NVP-BE235 will both block

mTORC1 which will lead to the induction of autophagy.

Addition of a suboptimal dose of metformin did not lower the rapamycin IC

⁠50

in ASPC-1cells (Fig. 6, Panel A). In contrast,

addi-tion of a suboptimal dose of metformin lowered the growth and rapamycin IC

⁠50

in BxPC-3cells from approximately 45nM–0.85nM

(8)

UNCORRECTED

PROOF

Fig. 4. Effects of Cisplatin and Vismodegib in the Absence and Presence of Suboptimal Doses of Metformin on the Proliferation of Three Pancreatic Cancer Cell Lines. ASPC-1 (Panel A), BxPC-3 (Panel B) and MIA-PaCa-2 (Panel C) were titrated with different concentrations of cisplatin in the absence (red squares) and

presence (blue triangles) of suboptimal doses of 250nM metformin. Arrows on the X-axis indicate where the IC⁠50s can be estimated. In Panel A, the two-tailed P value

is less than 0.0001 between cisplatin and cisplatin and metformin-treated ASPC-1cells and considered to be extremely statistically significant. In Panel B, the two-tailed

P value is less than 0.0001 between the cisplatin and the cisplatin and metformin-treated BxPC-1cells and considered to be extremely statistically significant. ASPC-1

(Panel D), BxPC-3 (Panel E) and MIA-PaCa-2 (Panel F) were titrated with different concentrations of vismodegib in the absence (red squares) and presence (blue trian-gles) of suboptimal doses of 250nM metformin. Arrows on the X-axis indicate where the IC⁠50s can be estimated. These experiments were repeated 3 times and similar

results were obtained. Adh.=adherent cells. (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.)

Fig. 5. Effects of Doxorubicin and Docetaxel in the Absence and Presence of Suboptimal Doses of Metformin on the Proliferation of Three Pancreatic Cancer Cell Lines. ASPC-1 (Panel A), BxPC-3 (Panel B) and MIA-PaCa-2 (Panel C) were titrated with different concentrations of doxorubicin in the absence (red squares) and

presence (blue triangles) of suboptimal doses of 250nM metformin. Arrows on the X-axis indicate where the IC⁠50s can be estimated. In Panel A, the two-tailed P value

is less than 0.0001 between doxorubicin and doxorubicin and metformin-treated ASPC-1cells and considered to be extremely statistically significant. In Panel C, the two-tailed P value is less than 0.0001 between the doxorubicin and the doxorubicin and metformin-treated MIA-PaCa-2cells are less than 0.0001 and considered to be extremely statistically significant. ASPC-1 (Panel D), BxPC-3 (Panel E) and MIA-PaCa-2 (Panel F) were titrated with different concentrations of docetaxel in the absence (red squares) and presence (blue triangles) of suboptimal doses of 250nM metformin. Arrows on the X-axis indicate where the IC⁠50s can be estimated. In Panel D, the

two-tailed P value is less than 0.0001 between the docetaxel and the docetaxel and metformin-treated ASPC-1cells and considered to be extremely statistically signif-icant. These experiments were repeated 3 times and similar results were obtained. Adh.=adherent cells. (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.)

(9)

UNCORRECTED

PROOF

Fig. 6. Effects of Rapamycin and NVP-BE235 in the Absence and Presence of Suboptimal Doses of Metformin on the Proliferation of Three Pancreatic Cancer Cell Lines. ASPC-1 (Panel A), BxPC-3 (Panel B) and MIA-PaCa-2 (Panel C) were titrated with different concentrations of rapamycin in the absence (red squares) and

presence (blue triangles) of suboptimal doses of 250nM metformin. Arrows on the X-axis indicate where the IC⁠50s can be estimated. In Panel B, the two-tailed P value

is less than 0.0001 between the rapamycin and the rapamycin and metformin-treated BxPC-1cells and considered to be extremely statistically significant. In Panel C, the two-tailed P value is less than 0.0001 between the rapamycin and the rapamycin and metformin-treated MIA-PaCa-2cells are less than 0.0001 and considered to be extremely statistically significant. ASPC-1 (Panel D), BxPC-3 (Panel E) and MIA-PaCa-2 (Panel F) were titrated with different concentrations of NVP-BE235 in the absence (red squares) and presence (blue triangles) of suboptimal doses of 250nM metformin. Arrows on the X-axis indicate where the IC⁠50s can be estimated. In

Panel E, the two-tailed P value is less than 0.0001 between the NVP-BE-235 and the NVP-BE235 and metformin-treated BxPC-3cells is less than 0.0001 and considered to be extremely statistically significant. These experiments were repeated 3 times and similar results were obtained. Adh.=adherent cells. (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.)

(approximately 53-fold) (Fig. 6, Panel B). Addition of a suboptimal dose of metformin lowered the rapamycin IC

⁠50

in MIA-PaCa-2cells

from approximately 60nM–5nM (about 12-fold) (Fig. 6, Panel C). These experiments demonstrated that suboptimal doses of

met-formin could lower the concentration of rapamycin required to reach the IC

⁠50

for two different PDAC lines, BxPC-3 and MIA-PaCa-2

lines but not in AsPC-1cells. Interestingly, BxPC-3 is reported to have wild-type (WT) KRAS, while ASPC-1 and MIA-PaCa-2cells have

mutant KRAS genes (Deer et al., 2010; Barretina et al., 2012; Hamidi et al., 2014).

Addition of a suboptimal dose of metformin did not lower the dual PI3K/mTORC NVP-BE235 inhibitor IC

⁠50

in ASPC-1cells (Fig. 6,

Panel D). Addition of a suboptimal dose of metformin did lower the NVP-BE235 IC

⁠50

in BxPC-3cells from 28 to 1.3nM, approximately

21.5-fold (Fig. 6, Panel E). Addition of a suboptimal dose of metformin did not lower the NVP-BE235 IC

⁠50

in MIA-PaCa-2cells (Fig. 6,

Panel F).

2.4. Effects of combining a suboptimal dose of metformin with the nutraceutical berberine and the anti-malarial drug chloroquine

Berberine is a natural product obtained from various berries and fruits (McCubrey et al., 2017a, 2017b, 2017e). It can have

multi-ple effects on cells, one is on the induction of NF-kappa B, another may be the induction of AMPK. The plant natural product berberine

has recently been shown to inhibit the growth of breast and pancreatic cancer cells (McCubrey et al., 2017e). Berberine may also

target and activate LKB1/AMPK (McCubrey et al., 2017a; Pan et al., 2017). Berberine may also target various types of CSCs (Lin et

al., 2017).

Chloroquine is a drug which is commonly used to eliminate malarial disease. The effects of treatment of PDAC cells with

subopti-mal doses of metformin and either berberine or chloroquine are presented in Fig. 7.

Addition of a suboptimal dose of metformin lowered the berberine IC

⁠50

from in ASPC-1cells from approximately 1800nM–30nM,

approximately 60-fold (Fig. 7, Panel A). Addition of a suboptimal dose of metformin lowered the growth and berberine IC

⁠50

in

BxPC-3cells from approximately 600nM–20nM (approximately 30-fold) (Fig. 6, Panel B). In contrast, addition of a suboptimal dose

of metformin did not lower the berberine IC

⁠50

in MIA-PaCa-2cells (Fig. 7, Panel C). These experiments demonstrated that suboptimal

doses of metformin could lower the concentration of berberine required to reach the IC

⁠50

for two different PDAC lines, ASPC-1 and

BxPC-3 lines but not in MIA-PaCa-2cells.

Addition of a suboptimal dose of metformin lowered the chloroquine IC

⁠50

in ASPC-1cells from 2μM to 0.35μM (approximately

5.7-fold) (Fig. 7, Panel D). Addition of a suboptimal dose of metformin lowered the chloroquine IC

⁠50

in BxPC-3cells from 10 to 0.15μM

approximately 67-fold (Fig. 7, Panel E). Addition of a suboptimal dose of metformin did lower the chloroquine IC

⁠50

in MIA-PaCa-2cells

from greater than 20μM to approximately 0.2μM (approximately 100-fold) (Fig. 7, Panel F). Thus, the effects of chloroquine in all

three PDAC lines can be enhanced by a suboptimal dose of metformin.

(10)

UNCORRECTED

PROOF

Fig. 7. Effects of Berberine and Chloroquine in the Absence and Presence of Suboptimal Doses of Metformin on the Proliferation of Three Pancreatic Cancer Cell Lines. ASPC-1 (Panel A), BxPC-3 (Panel B) and MIA-PaCa-2 (Panel C) were titrated with different concentrations of berberine in the absence (red squares) and

presence (blue triangles) of suboptimal doses of 250nM metformin. Arrows on the X-axis indicate where the IC⁠50s can be estimated. In Panel A, the two-tailed P value is

less than 0.0001 between berberine and berberine and metformin-treated ASPC-1cells and considered to be extremely statistically significant. In Panel B, the two-tailed

P value is less than 0.0001 between the berberine and the berberine and metformin-treated BxPC-1cells and considered to be extremely statistically significant. ASPC-1

(Panel D), BxPC-3 (Panel E) and MIA-PaCa-2 (Panel F) were titrated with different concentrations of chloroquine in the absence (red squares) and presence (blue trian-gles) of suboptimal doses of 250nM metformin. Arrows on the X-axis indicate where the IC⁠50s can be estimated. In Panel D, the two-tailed P value is less than 0.0001

between the chloroquine and the chloroquine and metformin-treated ASPC-1cells and considered to be extremely statistically significant. In Panel E, the two-tailed P value is less than 0.0001 between the chloroquine and the chloroquine and metformin-treated BxPC-3cells is less than 0.0001 and considered to be extremely statisti-cally significant. These experiments were repeated 3 times and similar results were obtained. Adh.=adherent cells. (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.)

3. Conclusions

Metformin is one of the most frequently prescribed drugs to treat type-II diabetes. It acts by inducing AMPK which in turn

sup-presses mTORC1 that results in the induction of autophagy. Metformin treatment by itself did not inhibit growth of PDAC cell lines

as much as 5-fluorouracil or gemcitabine which are used to treat PDAC patients. However, a suboptimal dose of metformin could

lower the IC

⁠50

s for various chemotherapeutic drugs. The most substantial effects were observed with 5-fluorouracil and gemcitabine

in all three PDACs examined. Metformin also enhanced the effects of cisplatin in the BxPC-3cell line. Gemcitabine, 5-fluorouracil

and cisplatin are all nucleoside analogues. The significance of these observations is being examined with other nucleoside analogues

used to treat PDAC. Metformin showed some effects on the effects on doxorubicin (a topoisomerase inhibitor) in AsPC-1, BxPC3 and

MIA-PaCa-2cells. In contrast, metformin did not appear to have effects at this suboptimal concentration when combined with

doc-etaxel (microtubule depolymerization blocker) in the PDAC lines examined. These two other chemotherapeutic drugs act by different

mechanisms than the nucleoside analogues.

Although, initially one might think that metformin, rapalogs and PI3K/Akt inhibitors might have similar effects, they have some

different properties which may lead to their combined usage might prove more effective. For example, while metformin suppresses

the proliferation of T-acute lymphoblastic leukemias (T-ALL) cell lines and primary T-ALL patient samples, metformin induced

au-tophagy and apoptosis and suppressed proliferation. The effects of metformin on control CD4

⁠+

T-lymphocytes were also examined.

Metformin was determined to be less toxic against CD4

⁠+

T-lymphocytes from healthy donors than against either the T-ALL cell lines

or primary patient cells. These studies documented the dephosphorylation of downstream targets of mTORC1. In addition, inhibition

of mRNA translation in metformin-treated T-ALL cells were seen. Importantly, a similar inhibition of translation was not observed

after rapamycin treatment. Intriguingly, in primary patient samples, metformin targeted the side population of T-ALL cell lines as well

as a putative leukemia initiating cell (LIC) subpopulation (CD34

⁠+

, CD7-, CD4-). Metformin exhibited anti-leukemic activity, which

supports further development of LKB1/AMPK pathway activators as potential clinical candidates for T-ALL and other cancer therapies

(Grimaldi et al., 2012).

AMPK is also important in BCR-ABL-induced chronic myeloid leukemia (CML) and AMPK pathway activators may prove useful

as combination drug therapy (with BCR-ABL inhibitors) for this disease (Martelli et al., 2012). AMPK can also be activated by

ra-pamycin (Habib, 2011). These studies also demonstrated that the effects of rara-pamycin and metformin are not always identical. In our

studies, we observed that suboptimal doses of metformin could enhance the effects of rapamycin or the PI3K/mTOR dual inhibitor

NVP-BEZ-235 in certain PDAC lines which indicate that while these inhibitors and metformin may have some similar downstream

targets, there may be others which upon suppression can eliminate cellular proliferation more effectively.

(11)

UNCORRECTED

PROOF

Metformin could also enhance the effects of the mTORC1 blocker rapamycin and the PI3K/mTOR inhibitor NVP-BE235 in both

the BxPC-3 and MIA-PaCa-2cells but not in the ASPC-1cells. BxPC-3cells are not reported to have mutant KRAS, but both ASPC-1 and

MIA-PaCa-2 both have mutant KRAS. All three cell lines have mutant KRAS but they differ in the presence of other key genetic

mu-tations. These studies document that different approaches to suppress the PI3K pathway and mTORC1 can be effective. That is, even

though rapamycin and NVP-BE-235 do act in part to inhibit mTORC, metformin can enhance their abilities to suppress cell growth in

at least two PDAC models.

Berberine is used for the treatment and prevention of various diseases world-wide now (McCubrey et al., 2017a, 2017b, 2017c,

2017d, 2017e; McCubrey and Cocco, 2017). Berberine are naturally present in many fruits and plants including: Berberis aetnensis C.

Presl., Berberis aristata, Berberis vulgaris, Coptis chinensis, Coptis japonica, Coptis rhizome, Hydrastis canadensis, Phellondendron amurense

and Tinosora cordifolia. Berberine is an isoquinoline quaternary alkaloid (a 5,6-dihydrodibenzo[a,g]quinolizinium derivative). The

po-tential beneficial effects of BBR are being investigated in at least 35 clinical trials (McCubrey et al., 2017a).

Berberine can modulate the expression of various genes that function in the regulation of cell growth including: BCL2, BCLXL,

PARP1, Beclin-1, TP53, p21

⁠Cip1

, MMP9 (Tillhon et al., 2012). Among the many effects of berberine which have been described it may

induce double strand DNA breaks and cell cycle arrest (Wang et al., 2012).

Berberines can form complexes with nucleic acids, such as DNA and RNA by binding the nitrogen atom at the 7-positon in the

alkaloid BBR skeleton. These complexes may interfere telomerases and topoisomerases (Gatto et al., 1996; Kim et al., 1998; Qin et al.,

2007; Bhowmik et al., 2012). Berberines may also affect mRNA transcription by interacting with the TATA-binding proteins and the

TATA-boxes contained in certain promoter regions of anti-apoptotic genes such as BCL2 and subsequently influence gene transcription

(Wang et al., 2011; Xiao et al., 2012).

Berberine can affect many different physiological processes as well as cancer and non-alcoholic fatty liver disease (Zhu et al., 2016;

McCubrey et al., 2017a, 2017b, 2017c, 2017d, 2017e). Treatment of ASPC-1 and BxPC-1cells with suboptimal dose of metformin

lowered the concentrations of berberine required to reach the IC

⁠50

. In contrast, the combined effects of berberine and metformin on

suppression of proliferation were not observed on MIA-PaCa-2cells. Berberine and other natural products that we consume in our diet

may have effects on the induction of reactive oxygen species which can have anti-cancer effects via stimulation of apoptosis and other

growth regulatory processes (Park et al., 2015; Cusimano et al., 2017).

The anti-malarial drug chloroquine is known to inhibit the induction of autophagy. Thus, in theory the addition of chloroquine

and metformin could have opposing effects. However, we observed that the combined treatment of a suboptimal dose of metformin

enhanced the effects of chloroquine on the suppression of growth in all three pancreatic cell lines examined. Likewise blockade of

MEK inhibitor induced autophagy has been shown to sensitize certain cancer cells to the PI3K/AKT inhibitor NVP-BKM120 (Ren et

al., 2016).

These three pancreatic cell lines differ in the frequency and presence of sphere forming cells in them upon growth under normal

culture conditions. There are often many sphere forming cells present in MIA-PaCa-2cells. Sphere forming cells have some

charac-teristics of CSCs. It will be interesting to determine whether the sphere forming cells from the various PDAC lines have different

sensitivities to metformin and various other drugs and nutraceuticals.

Our studies document that while some drug combinations with metformin will result in a lowering of the drug concentration

re-quired to meet the IC

⁠50

in all three cell lines examined, some metformin and drug combinations do not lead to a lowering of the drug

concentration required to reach the IC

⁠50

s. These results are important as in cancer patients, there is heterogeneity in pancreatic cancer

cell lines in terms to their responses to certain drugs and drug combinations. Thus, we should not expect that some drug combinations

will work with all cancer patients.

4. Materials and methods

4.1. Cell lines

The MIA-PaCa-2 PDAC (ATCC

⁠®

CRM-CRL-1420™) carcinoma cell line was derived from a 65-year old Caucasian male (Deer et al.,

2010). MIA-PaCa-2cells have the R248W TP53 GOF mutation. The R248W TP53 mutation present in MIA-PaCa-2cells is a missense

point mutation in the central DNA binding domain which abrogates its DNA contact (Liu et al., 2010). This TP53 mutation results

in a TP53 protein that is unable to bind specifically TP53 target sequences in TP53-responsive genes and results in loss of its tumor

suppressor properties (Song et al., 2007; Solomon et al., 2012). MIA-PaCa-2cells also have an activating mutation at KRAS (G12C)

and they have elevated PI3K/AKT pathway activity.

The ASPC-1cell line (ATCC

⁠®

CRL-1682™), an adenocarcinoma, was derived from a 62-year old female (Deer et al., 2010). It has

mutant KRAS and TP53 genes.

The BxPC-3cell line (ATCC

⁠®

CRL-1687™), an adenocarcinoma, was derived from a 61-year-old woman (Deer et al., 2010).

Inter-estingly, the BxPC-3cells express WT-KRAS (Barretina et al., 2012) but it has mutant TP53 genes.

4.2. Tissue culture and treatment of cells with signal transduction inhibitors and doxorubicin

Tissue culture medium was obtained from Invitrogen (Carlsbad, CA, USA). Nutraceuticals, chemotherapeutic drugs, signal

trans-duction inhibitors were purchased from either Selleckchem (Houston, TX USA) or Sigma-Aldrich (Saint Louis, MO, USA).

(12)

MIA-PaCa-UNCORRECTED

PROOF

2 and MCF-7cells were titrated with the different nutraceuticals, signal transduction inhibitors, chemotherapeutic and other drugs

as described (Chappell et al., 2012; Abrams et al., 2017; Steelman et al., 2018). Statistical analysis was performed using GraphPad

Prism.

Conflicts of interest

The authors declare that they have no conflicts of interest with publication of this manuscript.

Acknowledgements

JAM, SLA and KL were supported in part by a grant from East Carolina University Grants (#111104 and #111110-668715-0000).

LC and SR were supported in part by grants from: Intesa San Paolo Foundation. ML and SC were supported in part by a Lega

Ital-iana per la Lotta contro i Tumori grant. DR and AG were supported in part from the Polish National Science Center grant number

UMO-2015/19/B/NZ1/00332.

References

Abrams, S.L., Ruvolo, P.P., Ruvolo, V.R., Ligresti, G., Martelli, A.M., Cocco, L., Ratti, S., Tafuri, A., Steelman, L.S., Candido, S., Libra, M., McCubrey, J.M., 2017. Tar-geting signaling and apoptotic pathways involved in chemotherapeutic drug-resistance of hematopoietic cells. Oncotarget 8, 76525–76557.

Adamska, A., Elaskalani, O., Emmanouilidi, A., Kim, M., Abdol Razak, N.B., Metharom, P., Falasca, M., 2017. Molecular and cellular mechanisms of chemoresistance in pancreatic cancer. Adv. Biol. Regul. https://doi.org/10.1016/j.jbior.2017.11.007, pii: S2212–4926(17)30176-8.

Alfarouk, K.O., Verduzco, D., Rauch, C., Muddathir, A.K., Adil, H.H., Elhassan, G.O., Ibrahim, M.E., David Polo Orozco, J., Cardone, R.A., Reshkin, S.J., Harguindey, S., 2014. Glycolysis, tumor metabolism, cancer growth and dissemination. A new pH-based etiopathogenic perspective and therapeutic approach to an old cancer question. Oncoscience 1, 777–802.

Amin, S., Mhango, G., Lin, J., Aronson, A., Wisnivesky, J., Boffetta, P., Lucas, A., 2016. Metformin improves survival in patients with pancreatic ductal adenocarcinoma and pre-existing diabetes: a propensity score analysis. Am. J. Gastroenterol. 111, 1350–1357.

Anderson, K.E., Juvin, V., Clark, J., Stephens, L.R., Hawkins, P.T., 2016. Investigating the effect of arachidonate supplementation on the phosphoinositide content of MCF10a breast epithelial cells. Adv. Biol. Regul. 62, 18–24.

Ando, H., Kawaai, K., Bonneau, B., Mikoshiba, K., 2017. Remodeling of Ca2+ signaling in cancer: regulation of inositol 1,4,5-trisphosphate receptors through onco-genes and tumor suppressors. Adv. Biol. Regul. https://doi.org/10.1016/j.jbior.2017.12.001, pii: S2212–4926(17)30188-4.

Arslan, A.A., Helzlsouer, K.J., Kooperberg, C., Shu, X.O., Steplowski, E., Bueno-de-Mesquita, H.B., Fuchs, C.S., Gross, M.D., Jacobs, E.J., Lacroix, A.Z., Petersen, G.M., Stolzenberg-Solomon, R.Z., Zheng, W., Albanes, D., Amundadottir, L., Bamlet, W.R., Barricarte, A., Bingham, S.A., Boeing, H., Boutron-Ruault, M.C., Buring, J.E., Chanock, S.J., Clipp, S., Gaziano, J.M., Giovannucci, E.L., Hankinson, S.E., Hartge, P., Hoover, R.N., Hunter, D.J., Hutchinson, A., Jacobs, K.B., Kraft, P., Lynch, S.M., Manjer, J., Manson, J.E., McTiernan, A., McWilliams, R.R., Mendelsohn, J.B., Michaud, D.S., Palli, D., Rohan, T.E., Slimani, N., Thomas, G., Tjønneland, A., Tobias, G.S., Trichopoulos, D., Virtamo, J., Wolpin, B.M., Yu, K., Zeleniuch-Jacquotte, A., Patel, A.V., 2010. Anthropometric measures, body mass index, and pancreatic cancer: a pooled analysis from the pancreatic cancer cohort consortium (PanScan). Arch. Intern. Med. 170, 791–802.

Balic, A., Sørensen, M.D., Trabulo, S.M., Sainz Jr., B., Cioffi, M., Vieira, C.R., Miranda-Lorenzo, I., Hidalgo, M., Kleeff, J., Erkan, M., Heeschen, C., 2014a. Chloroquine targets pancreatic cancer stem cells via inhibition of CXCR4 and hedgehog signaling. Mol. Cancer Ther. 13, 1758–1771.

Balic, A., Sorensen, M.D., Heeschen, C., 2014b. Old drugs for new purposes-chloroquine targets metastatic pancreatic cancer stem cells & their microenvironment. Cancer Cell Microenviron. 1, e227.

Banfic, H., Crljen, V., Lukinovic-Skudar, V., Dembitz, V., Lalic, H., Bedalov, A., Visnjic, D., 2016. Inositol pyrophosphates modulate cell cycle independently of alter-ation in telomere length. Adv. Biol. Regul. 60, 22–28.

Bao, B., Wang, Z., Ali, S., Ahmad, A., Azmi, A.S., Sarkar, S.H., Banerjee, S., Kong, D., Li, Y., Thakur, S., Sarkar, F.H., 2012. Metformin inhibits cell proliferation, migra-tion and invasion by attenuating CSC funcmigra-tion mediated by deregulating miRNAs in pancreatic cancer cells. Cancer Prev. Res. 5, 355–364.

Bao, B., Azmi, A.S., Ali, S., Zaiem, F., Sarkar, F.H., 2014. Metformin may function as anti-cancer agent via targeting cancer stem cells: the potential biological signifi-cance of tumor-associated miRNAs in breast and pancreatic signifi-cancers. Ann. Transl. Med. 2, 59.

Barretina, J., Caponigro, G., Stransky, N., Venkatesan, K., Margolin, A.A., Kim, S., Wilson, C.J., Lehá, r.J., Kryukov, G.V., Sonkin, D., Reddy, A., Liu, M., Murray, L., Berger, M.F., Monahan, J.E., Morais, P., Meltzer, J., Korejwa, A., Jané-Valbuena, J., Mapa, F.A., Thibault, J., Bric-Furlong, E., Raman, P., Shipway, A., Engels, I.H., Cheng, J., Yu, G.K., Yu, J., Aspesi Jr., P., de Silva, M., Jagtap, K., Jones, M.D., Wang, L., Hatton, C., Palescandolo, E., Gupta, S., Mahan, S., Sougnez, C., Onofrio, R.C., Liefeld, T., MacConaill, L., Winckler, W., Reich, M., Li, N., Mesirov, J.P., Gabriel, S.B., Getz, G., Ardlie, K., Chan, V., Myer, V.E., Weber, B.L., Porter, J., War-muth, M., Finan, P., Harris, J.L., Meyerson, M., Golub, T.R., Morrissey, M.P., Sellers, W.R., Schlegel, R., Garraway, L.A., 2012. The cancer cell line encyclopedia enables predictive modelling of anticancer drug sensitivity. Nature 483, 603–607.

Bhowmik, D., Das, S., Hossain, M., Haq, L., Kumar, G.S., 2012. Biophysical characterization of the strong stabilization of the RNA triplex poly(U)·poly(A)∗poly(U) by 9-O-(-amino) alkyl ether berberine analogs. PLoS One 7, e37939.

Bosetti, C., Rosato, V., Li, D., Silverman, D., Petersen, G.M., Bracci, P.M., Neale, R.E., Muscat, J., Anderson, K., Gallinger, S., Olson, S.H., Miller, A.B., Bas Bueno-de-Mesquita, H., Scelo, G., Janout, V., Holcatova, I., Lagiou, P., Serraino, D., Lucenteforte, E., Fabianova, E., Ghadirian, P., Baghurst, P.A., Zatonski, W., Foretova, L., Fontham, E., Bamlet, W.R., Holly, E.A., Negri, E., Hassan, M., Prizment, A., Cotterchio, M., Cleary, S., Kurtz, R.C., Maisonneuve, P., Trichopoulos, D., Polesel, J., Duell, E.J., Boffetta, P., La Vecchia, C., 2014. Diabetes, antidiabetic medications, and pancreatic cancer risk: an analysis from the international pancre-atic cancer case-control consortium. Ann. Oncol. 25, 2065–2072.

Bressanin, D., Evangelisti, C., Ricci, F., Tabellini, G., Chiarini, F., Tazzari, P.L., Melchionda, F., Buontempo, F., Pagliaro, P., Pession, A., McCubrey, J.A., Martelli, A.M., 2012. Harnessing the PI3K/Akt/mTOR pathway in T-cell acute lymphoblastic leukemia: eliminating activity by targeting at different levels. Oncotarget 3, 811–823. Campa, C.C., Hirsch, E., 2017. Rab11 and phosphoinositides: a synergy of signal transducers in the control of vesicular trafficking. Adv. Biol. Regul. 63, 132–139. Carman, G.M., Han, G.S., 2017. Phosphatidate phosphatase regulates membrane phospholipid synthesis via phosphatidylserine synthase. Adv. Biol. Regul. 16, pii:

S2212–4926(17)30129X.

Catenacci, D.V., Junttila, M.R., Karrison, T., Bahary, N., Horiba, M.N., Nattam, S.R., Marsh, R., Wallace, J., Kozloff, M., Rajdev, L., Cohen, D., Wade, J., Sleckman, B., Lenz, H.J., Stiff, P., Kumar, P., Xu, P., Henderson, L., Takebe, N., Salgia, R., Wang, X., Stadler, W.M., de Sauvage, F.J., Kindler, H.L., 2015. Randomized phase Ib/ II study of gemcitabine plus placebo or vismodegib, a hedgehog pathway inhibitor, in patients with metastatic pancreatic cancer. J. Clin. Oncol. 33, 4284–4292. Cerullo, M., Gani, F., Chen, S.Y., Canner, J., Pawlik, T.M., 2017. Metformin use Is associated with improved survival in patients undergoing resection for pancreatic

cancer. J. Gastrointest. Surg. 20, 1572–1580.

Cervello, M., Augello, G., Cusimano, A., Emma, M.R., Balasus, D., Azzolina, A., McCubrey, J.A., Montalto, G., 2017. Pivotal roles of glycogen synthase-3 in hepatocel-lular carcinoma. Adv. Biol. Regul. 65, 59–76.

(13)

UNCORRECTED

PROOF

Chaiteerakij, R., Petersen, G.M., Bamlet, W.R., Chaffee, K.G., Zhen, D.B., Burch, P.A., Leof, E.R., Roberts, L.R., Oberg, A.L., 2016. Metformin use and survival of patients with pancreatic cancer: a cautionary lesson. J. Clin. Oncol. 34, 1898–1904.

Chappell, W.H., Abrams, S.L., Franklin, R.A., LaHair, M.M., Montalto, G., Cervello, M., Martelli, A.M., Nicoletti, F., Candido, S., Libra, M., Polesel, J., Talamini, R., Milella, M., Tafuri, A., Steelman, L.S., McCubrey, J.A., 2012. Ectopic NGAL expression can alter sensitivity of breast cancer cells to EGFR, Bcl-2, CaM-K inhibitors and the plant natural product berberine. Cell Cycle 11, 4447–4461.

Chappell, W.H., Abrams, S.L., Lertpiriyapong, K., Fitzgerald, T.L., Martelli, A.M., Cocco, L., Rakus, D., Gizak, A., Terrian, D., Steelman, L.S., McCubrey, J.A., 2016. Novel roles of androgen receptor, epidermal growth factor receptor, TP53, regulatory RNAs, NF-kappa-B, chromosomal translocations, neutrophil associated gelatinase, and matrix metalloproteinase-9 in prostate cancer and prostate cancer stem cells. Adv. Biol. Regul. 60, 64–87.

Cioce, M., Valerio, M., Casadei, L., Pulito, C., Sacconi, A., Mori, F., Biagioni, F., Manetti, C., Muti, P., Strano, S., Blandino, G., 2014. Metformin-induced metabolic reprogramming of chemoresistant ALDHbright breast cancer cells. Oncotarget 5, 4129–4143.

Coant, N., Sakamoto, W., Mao, C., Hannun, Y.A., 2017. Ceramidases, roles in sphingolipid metabolism and in health and disease. Adv. Biol. Regul. 63, 122–131. Cocco, L., Manzoli, L., Faenza, I., Ramazzotti, G., Yang, Y.R., McCubrey, J.A., Suh, P.G., Follo, M.Y., 2016. Modulation of nuclear PI-PLCbeta1 during cell

differentia-tion. Adv. Biol. Regul. 60, 1–5.

Cusimano, A., Balasus, D., Azzolina, A., Augello, G., Emma, M.R., Di Sano, C., Gramignoli, R., Strom, S.C., McCubrey, J.A., Montalto, G., Cervello, M., 2017. Oleocan-thal exerts antitumor effects on human liver and colon cancer cells through ROS generation. Int. J. Oncol. 51, 533–544.

Cuyas, E., Corominas-Faja, B., Menendez, J.A., 2014. The nutritional phenome of EMT-induced cancer stem-like cells. Oncotarget 5, 3970–3982.

Deer, E.L., González-Hernández, J., Coursen, J.D., Shea, J.E., Ngatia, J., Scaife, C.L., Firpo, M.A., Mulvihill, S.J., 2010. Phenotype and genotype of pancreatic cancer cell lines. Pancreas 39, 425–435.

Dembinski, J.L., Krauss, S., 2009. Characterization and functional analysis of a slow cycling stem cell-like subpopulation in pancreas adenocarcinoma. Clin. Exp. Metas-tasis 26, 611–623.

Ebenezer, D.L., Fu, P., Suryadevara, V., Zhao, Y., Natarajan, V., 2017. Epigenetic regulation of pro-inflammatory cytokine secretion by sphingosine 1-phosphate (S1P) in acute lung injury: role of S1P lyase. Adv. Biol. Regul. 63, 156–166.

Erneux, C., Ghosh, S., Koenig, S., 2016. Inositol (1,4,5)P3 3-kinase isoenzymes: catalytic properties and importance of targeting to F-actin to understand function. Adv. Biol. Regul. 60, 135–143.

Falasca, M., Ferro, R., 2016. Role of the lysophosphatidylinositol/GPR55 axis in cancer. Adv. Biol. Regul. 60, 88–93.

Fields, A.P., Justilien, V., Murray, N.R., 2016. The chromosome 3q26 OncCassette: a multigenic driver of human cancer. Adv. Biol. Regul. 60, 47–63.

Fitzgerald, T.L., McCubrey, J.A., 2014. Pancreatic cancer stem cells: association with cell surface markers, prognosis, resistance, metastasis and treatment. Adv. Biol. Regul. 56, 45–50.

Fitzgerald, T.L., Lertpiriyapong, K., Cocco, L., Martelli, A.M., Libra, M., Candido, S., Montalto, G., Cervello, M., Steelman, L., Abrams, S.L., McCubrey, J.A., 2015. Roles of EGFR and KRAS and their downstream signaling pathways in pancreatic cancer and pancreatic cancer stem cells. Adv. Biol. Regul. 59, 65–81.

Frid, A., Sterner, G.N., Löndahl, M., Wiklander, C., Cato, A., Vinge, E., Andersson, A., 2010. Novel assay of metformin levels in patients with type 2 diabetes and varying levels of renal function: clinical recommendations. Diabetes Care 33, 1291–1293.

Frouws, M.A., Mulde, r.B.G., Bastiaannet, E., Zanders, M.M., van Herk-Sukel, M.P., de Leede, E.M., Bonsing, B.A., Mieog, J.S., Van de Velde, C.J., Liefers, G.J., 2017. No association between metformin use and survival in patients with pancreatic cancer: an observational cohort study. Medicine 96, 6229.

Fujisawa, T., 2017. Therapeutic application of apoptosis signal-regulating kinase 1 inhibitors. Adv. Biol. Regul. 66, 85–90.

Gatto, B., Sanders, M.M., Yu, C., Wu, H.Y., Makhey, D., LaVoie, E.J., Liu, L.F., 1996. Identification of topoisomerase I as the cytotoxic target of the protoberberine alkaloid coralyne. Cancer Res. 56, 2795–2800.

Geck, R.C., Toker, A., 2016. Nonessential amino acid metabolism in breast cancer. Adv. Biol. Regul. 62, 11–17. Geffken, K., Spiegel, S., 2017. Sphingosine kinase 1 in breast cancer. Adv. Biol. Regul.pii: S2212-4926(17) 30164–1.

Ghim, J., Chelakkot, C., Bae, Y.S., Suh, P.G., Ryu, S.H., 2016. Accumulating insights into the role of phospholipase D2 in human diseases. Adv. Biol. Regul. 61, 42–46. Giudici, M.L., Clarke, J.H., Irvine, R.F., 2016. Phosphatidylinositol 5-phosphate 4-kinase γ (PI5P4Kγ), a lipid signalling enigma. Adv. Biol. Regul. 61, 47–50. Godlewski, J., Bronisz, A., Nowicki, M.O., Chiocca, E.A., Lawler, S., 2010. microRNA-451: a conditional switch controlling glioma cell proliferation and migration. Cell

Cycle 9, 2742–2748.

Gowda, C., Soliman, M., Kapadia, M., Ding, Y., Payne, K., Dovat, S., 2017a. Casein kinase II (CK2), glycogen synthase Kinase-3 (GSK-3) and Ikaros mediated regulation of leukemia. Adv. Biol. Regul. 65, 16–25.

Gowda, C., Song, C., Kapadia, M., Payne, J.L., Hu, T., Ding, Y., Dovat, S., 2017b. Regulation of cellular proliferation in acute lymphoblastic leukemia by Casein Kinase II (CK2) and Ikaros. Adv. Biol. Regul. 63, 71–80.

Grimaldi, C., Chiarini, F., Tabellini, G., Ricci, F., Tazzari, P.L., Battistelli, M., Falcieri, E., Bortul, R., Melchionda, F., Iacobucci, I., Pagliaro, P., Martinelli, G., Pession, A., Barata, J.T., McCubrey, J.A., Martelli, A.M., 2012. AMP-dependent kinase/mammalian target of rapamycin complex 1 signaling in T-cell acute lymphoblastic leukemia: therapeutic implications. Leukemia 26, 91–100.

Guo, X., Namekata, K., Kimura, A., Harada, C., Harada, T., 2017. ASK1 in neurodegeneration. Adv. Biol. Regul. 66, 63–71. Habib, S.L., 2011. Mechanism of activation of AMPK and upregulation of OGG1 by rapamycin in cancer cells. Oncotarget 2, 958–959.

Hamidi, H., Lu, M., Chau, K., Anderson, L., Fejzo, M., Ginther, C., Linnartz, R., Zubel, A., Slamon, D.J., Finn, R.S., 2014. KRAS mutational subtype and copy number predict in vitro response of human pancreatic cancer cell lines to MEK inhibition. Br. J. Cancer 111, 1788–1801.

Hatch, A.J., Odom, A.R., York, J.D., 2017. Inositol phosphate multikinase dependent transcriptional control. Adv. Biol. Regul. 64, 9–19.

Hermida, M.A., Dinesh Kumar, J., Leslie, N.R., 2017. GSK3 and its interactions with the PI3K/AKT/mTOR signalling network. Adv. Biol. Regul. 65, 5–15.

Iodice, S., Gandini, S., Maisonneuve, P., Lowenfels, A.B., 2008. Tobacco and the risk of pancreatic cancer: a review and meta-analysis. Langenbeck's Arch. Surg. 393, 535–545.

Jang, H.J., Suh, P.G., Lee, Y.J., Shin, K.J., Cocco, L., Chae, Y.C., 2017. PLCγ1: potential arbitrator of cancer progression. Adv. Biol. Regul.pii: S2212-4926(17) 30167–7. Jansen, M., Ten Klooster, J.P., Offerhaus, G.J., Clevers, H., 2009. LKB1 and AMPK family signaling: the intimate link between cell polarity and energy metabolism.

Physiol. Rev. 89, 777–798.

Janzer, A., German, N.J., Gonzalez-Herrera, K.N., Asara, J.M., et al., 2014. Metformin and phenformin deplete tricarboxylic acid cycle and glycolytic intermediats dur-ing cell transformation and NTPs in cancer stem cells. Proc. Natl. Acad. Sci. U.S.A. 111, 10574–10579.

Jhanwar-Uniyal, M., Amin, A.G., Cooper, J.B., Das, K., Schmidt, M.H., Murali, R., 2017. Discrete signaling mechanisms of mTORC1 and mTORC2: connected yet apart in cellular and molecular aspects. Adv. Biol. Regul. 64, 39–48.

Jiao, L., Berrington de Gonzalez, A., Hartge, P., Pfeiffer, R.M., Park, Y., Freedman, D.M., et al., 2010. Body mass index, effect modifiers, and risk of pancreatic cancer: a pooled study of seven prospective cohorts. Cancer Causes Control 21, 1305–1314.

Jung, D.E., Wen, J., Oh, T., Song, S.Y., 2011. Differentially expressed microRNAs in pancreatic cancer stem cells. Pancreas 40, 1180–1187.

Kang, D.S., Yang, Y.R., Lee, C., Kim, S., Ryu, S.H., Suh, P.G., 2016. Roles of phosphoinositide-specific phospholipase Cγ1 in brain development. Adv. Biol. Regul. 60, 167–173.

Klein, A.P., 2012. Genetic susceptibility to pancreatic cancer. Mol. Carcinog. 51, 14–24.

Kim, S.A., Kwon, Y., Kim, J.H., Muller, M.T., Chung, I.K., 1998. Induction of topoisomerase II mediated DNA cleavage by a protoberberine alkaloid, berberrubine. Biochemistry 37, 16316–16324.

Lee, H., Park, H.J., Park, C.S., et al., 2014. Response of breast cancer cells and cancer stem cells to metformin and hyperthermia alone or combined. PLoS One 9, e87979.

Leibovitch, M., Topisirovic, I., 2017. Dysregulation of mRNA translation and energy metabolism in cancer. Adv. Biol. Regul. https://doi.org/10.1016/j.jbior.2017.11. 001, pii: S2212–4926(17)30170-7.

Figure

Updating...

References

Updating...

Related subjects :